Publications

Estimation of the noise level function for color images using mathematical morphology and non-parametric statistics

By Baptiste Esteban, Guillaume Tochon, Edwin Carlinet, Didier Verna

2022-04-08

In Proceedings of the 26th international conference on pattern recognition

Abstract

Noise level information is crucial for many image processing tasks, such as image denoising. To estimate it, it is necessary to find homegeneous areas within the image which contain only noise. Rank-based methods have proven to be efficient to achieve such a task. In the past, we proposed a method to estimate the noise level function (NLF) of grayscale images using the tree of shapes (ToS). This method, relying on the connected components extracted from the ToS computed on the noisy image, had the advantage of being adapted to the image content, which is not the case when using square blocks, but is still restricted to grayscale images. In this paper, we extend our ToS-based method to color images. Unlike grayscale images, the pixel values in multivariate images do not have a natural order relationship, which is a well-known issue when working with mathematical morphology and rank statistics. We propose to use the multivariate ToS to retrieve homogeneous regions. We derive an order relationship for the multivariate pixel values thanks to a complete lattice learning strategy and use it to compute the rank statistics. The obtained multivariate NLF is composed of one NLF per channel. The performance of the proposed method is compared with the one obtained using square blocks, and validates the soundness of the multivariate ToS structure for this task.

Continue reading

A benchmark of named entity recognition approaches in historical documents

By Nathalie Abadie, Edwin Carlinet, Joseph Chazalon, Bertrand Duménieu

2022-04-07

In Proceedings of the 15th IAPR international workshop on document analysis system

Abstract

Named entity recognition (NER) is a necessary step in many pipelines targeting historical documents. Indeed, such natural language processing techniques identify which class each text token belongs to, e.g. “person name”, “location”, “number”. Introducing a new public dataset built from 19th century French directories, we first assess how noisy modern, off-the-shelf OCR are. Then, we compare modern CNN- and Transformer-based NER techniques which can be reasonably used in the context of historical document analysis. We measure their requirements in terms of training data, the effects of OCR noise on their performance, and show how Transformer-based NER can benefit from unsupervised pre-training and supervised fine-tuning on noisy data. Results can be reproduced using resources available at https://github.com/soduco/paper-ner-bench-das22 and https://zenodo.org/record/6394464

Continue reading

Learning grayscale mathematical morphology with smooth morphological layers

By Romain Hermary, Guillaume Tochon, Élodie Puybareau, Alexandre Kirszenberg, Jesús Angulo

2022-04-04

In Journal of Mathematical Imaging and Vision

Abstract

The integration of mathematical morphology operations within convolutional neural network architectures has received an increasing attention lately. However, replacing standard convolution layers by morphological layers performing erosions or dilations is particularly challenging because the min and max operations are not differentiable. P-convolution layers were proposed as a possible solution to this issue since they can act as smooth differentiable approximation of min and max operations, yielding pseudo-dilation or pseudo-erosion layers. In a recent work, we proposed two novel morphological layers based on the same principle as the p-convolution, while circumventing its principal drawbacks, and showcased their capacity to efficiently learn grayscale morphological operators while raising several edge cases. In this work, we complete those previous results by thoroughly analyzing the behavior of the proposed layers and by investigating and settling the reported edge cases. We also demonstrate the compatibility of one of the proposed morphological layers with binary morphological frameworks.

Continue reading

Is it really easy to detect sybil attacks in c-ITS environments: A position paper

By Badis Hammi, Yacine Mohamed Idir, Sherali Zeadally, Rida Khatoun, Jamel Nebhen

2022-04-01

In IEEE Transactions on Intelligent Transportation Systems

Abstract

In the context of current smart cities, Cooperative Intelligent Transportation Systems (C-ITS) represent one of the main use case scenarios that aim to improve peoples? daily lives. Thus, during the last few years, numerous standards have been adopted to regulate such networks. Within a C-ITS, a large number of messages are exchanged continuously in order to ensure that the different applications operate efficiently. However, these networks can be the target of numerous attacks. The sybil attack is among the most dangerous ones. In a sybil attack, an attacker creates multiple identities and then disguises as several fake stations in order to interfere with the normal operations of the system or profit from provided services. We analyze recently proposed sybil detection approaches regarding their compliance with the current C-ITS standards as well as their evaluation methods. We provide several recommendations such as network and attack models as well as an urban and highway datasets that can be considered in future research in sybil attack detection.

Continue reading

Qu’est-ce que mon GNN capture vraiment ? Exploration des représentations internes d’un GNN

By Luca Veyrin-Forrer, Ataollah Kamal, Stefan Duffner, Marc Plantevit, Céline Robardet

2022-03-24

In Extraction et gestion des connaissances, EGC 2022, blois, france, 24 au 28 janvier 2022

Abstract

While existing GNN’s explanation methods explain the decision by studying the output layer, we propose a method that analyzes the hidden layers to identify the neurons that are co-activated for a class. We associate to them a graph.

Continue reading

Local intensity order transformation for robust curvilinear object segmentation

By Tianyi Shi, Nicolas Boutry, Yongchao Xu, Thierry Géraud

2022-03-22

In IEEE Transactions on Image Processing

Abstract

Segmentation of curvilinear structures is important in many applications, such as retinal blood vessel segmentation for early detection of vessel diseases and pavement crack segmentation for road condition evaluation and maintenance. Currently, deep learning-based methods have achieved impressive performance on these tasks. Yet, most of them mainly focus on finding powerful deep architectures but ignore capturing the inherent curvilinear structure feature (e.g., the curvilinear structure is darker than the context) for a more robust representation. In consequence, the performance usually drops a lot on cross-datasets, which poses great challenges in practice. In this paper, we aim to improve the generalizability by introducing a novel local intensity order transformation (LIOT). Specifically, we transfer a gray-scale image into a contrast- invariant four-channel image based on the intensity order between each pixel and its nearby pixels along with the four (horizontal and vertical) directions. This results in a representation that preserves the inherent characteristic of the curvilinear structure while being robust to contrast changes. Cross-dataset evaluation on three retinal blood vessel segmentation datasets demonstrates that LIOT improves the generalizability of some state-of-the-art methods. Additionally, the cross-dataset evaluation between retinal blood vessel segmentation and pavement crack segmentation shows that LIOT is able to preserve the inherent characteristic of curvilinear structure with large appearance gaps. An implementation of the proposed method is available at https://github.com/TY-Shi/LIOT.

Continue reading

Electricity price forecasting on the day-ahead market using machine learning

By Léonard Tschora, Erwan Pierre, Marc Plantevit, Céline Robardet

2022-03-10

In Applied Energy

Abstract

The price of electricity on the European market is very volatile. This is due both to its mode of production by different sources, each with its own constraints (volume of production, dependence on the weather, or production inertia), and by the difficulty of its storage. Being able to predict the prices of the next day is an important issue, to allow the development of intelligent uses of electricity. In this article, we investigate the capabilities of different machine learning techniques to accurately predict electricity prices. Specifically, we extend current state-of-the-art approaches by considering previously unused predictive features such as price histories of neighboring countries. We show that these features significantly improve the quality of forecasts, even in the current period when sudden changes are occurring. We also develop an analysis of the contribution of the different features in model prediction using Shap values, in order to shed light on how models make their prediction and to build user confidence in models.

Continue reading

Max-tree computation on GPUs

By Nicolas Blin, Edwin Carlinet, Florian Lemaitre, Lionel Lacassagne, Thierry Géraud

2022-03-09

In IEEE Transactions on Parallel and Distributed Systems

Abstract

In Mathematical Morphology, the max-tree is a region-based representation that encodes the inclusion relationship of the threshold sets of an image. This tree has been proven useful in numerous image processing applications. For the last decade, works have been led to improve the building time of this structure; mixing algorithmic optimizations, parallel and distributed computing. Nevertheless, there is still no algorithm that takes benefit from the computing power of the massively parallel architectures. In this work, we propose the first GPU algorithm to compute the max-tree. The proposed approach leads to significant speed-ups, and is up to one order of magnitude faster than the current State-of-the-Art parallel CPU algorithms. This work paves the way for a max-tree integration in image processing GPU pipelines and real-time image processing based on Mathematical Morphology. It is also a foundation for porting other image representations from Mathematical Morphology on GPUs.

Continue reading

Anomaly detection on static and dynamic graphs using graph convolutional neural networks

By Amani Abou Rida, Rabih Amhaz, Pierre Parrend

2022-03-01

In Robotics and AI for cybersecurity and critical infrastructure in smart cities

Abstract

Anomalies represent rare observations that vary significantly from others. Anomaly detection intended to discover these rare observations has the power to prevent detrimental events, such as financial fraud, network intrusion, and social spam. However, conventional anomaly detection methods cannot handle this problem well because of the complexity of graph data (e.g., irregular structures, relational dependencies, node/edge types/attributes/directions/multiplicities/weights, large scale, etc.) [1]. Thanks to the rise of deep learning in solving these limitations, graph anomaly detection with deep learning has obtained an increasing attention from many scientists recently. However, while deep learning can capture unseen patterns of multi-dimensional Euclidean data, there is a huge number of applications where data are represented in the form of graphs. Graphs have been used to represent the structural relational information, which raises the graph anomaly detection problem - identifying anomalous graph objects (i.e., vertex, edges, sub-graphs, and change detection). These graphs can be constructed as a static graph, or a dynamic graph based on the availability of timestamp. Recent years have observed a huge efforts on static graphs, among which Graph Convolutional Network (GCN) has appeared as a useful class of models. A challenge today is to detect anomalies with dynamic structures. In this chapter, we aim at providing methods used for detecting anomalies in static and dynamic graphs using graph analysis, graph embedding, and graph convolutional neural networks. For static graphs we categorize these methods according to plain and attribute static graphs. For dynamic graphs we categorize existing methods according to the type of anomalies that they can detect. Moreover, we focus on the challenges in this research area and discuss the strengths and weaknesses of various methods in each category. Finally, we provide open challenges for graph anomaly detection using graph convolutional neural networks on dynamic graphs.

Continue reading

ETAP: Experimental typesetting algorithms platform

By Didier Verna

2022-03-01

In ELS 2022, the 15th european lisp symposium

Abstract

We present the early development stages of ETAP, a platform for experimenting with typesetting algorithms. The purpose of this platform is twofold: while its primary objective is to provide building blocks for quickly and easily designing and testing new algorithms (or variations on existing ones), it can also be used as an interactive, real time demonstrator for many features of digital typography, such as kerning, hyphenation, or ligaturing.

Continue reading

Search

Tags